At first glance the concept of “leisure” – comprising social, recreational, and entertainment
activities – is apparently well-understood. Numerous scholars have noted, however, that
defining leisure is not at all as straightforward as might be initially assumed (see, e.g., Howe and Rancourt, 1990). This section first reviews and critiques several definitions of leisure. It then points to one key source of difficulty in defining it – the fact that the boundaries between leisure and other types of activities are not crisp – and discusses three ways in which this is true. One means of further defining an object is to classify the various forms in which it can be manifested, and also in this section we review several classification schemes for leisure activities that have been previously offered.

2.1 Definitions of Leisure
The literature contains a number of definitions of leisure. For example, the 130 Australian
adolescents studied by Passmore and French (2001) indicated that freedom of choice and
enjoyability were crucial to an activity being considered leisure. Similarly, Tinsley, et al. (1993, p. 447) define four necessary characteristics for a leisure experience to occur: “The individual must perceive the activity as (a) freely chosen, (b) intrinsically satisfying, (c) optimally arousing, and (d) requiring a sense of commitment.” But clearly at least the latter three characteristics can apply to subsistence and maintenance activities as well as leisure, and even the first characteristic, freedom of choice, can apply to numerous tasks within an individual’s job or to certain aspects of maintenance activities. Conversely, it seems rather strict not to consider an activity such as accompanying a spouse to a ball game to be leisure if the individual does not entirely freely choose it, or is not fully “committed” to it or “aroused” by it (see, e.g., Kelly, 1978).
Meurs and Kalfs (2000, p. 128) define “leisure time” as “all the time a person does not devote to ensuring their [sic] future welfare in a broad sense.” They indicate that this definition thus
excludes activities associated with generating income, running a household, and maintaining
physical well-being. They further define “leisure travel” as “all journeys not specifically made with the purpose of providing for the person’s future welfare or even for sustaining a normal life.” In other words, “there is no future penalty for not making these journeys.” Yet these definitions also seem restrictive. Leisure activities should certainly be considered essential to one’s psychological welfare, i.e. welfare “in a broad sense”, with a corresponding psychological penalty for their complete neglect. And the exclusion of activities that support physical wellbeing would eliminate a large category of recreational activities, such as participatory sports or exercise, that would normally be classified as leisure.
Interestingly, although they can be more readily deferred or “compressed” than can subsistence or maintenance activities, leisure activities are seemingly less readily transferred than the other two types.4 Work and maintenance activities are considered essential to the individual’s physical well-being (although these activities can also make an important contribution to one’s psychological well-being). As such, an individual can receive similar physical benefits from outsourcing many 5 of the latter two types of activities to other individuals (e.g. by marrying a person who supports the household financially, or by hiring domestic help). In contrast, since the main contribution of leisure activities is to psychological well-being (although recreational activities can also support the physical dimension, as mentioned above), the individual does not benefit by outsourcing leisure to others 6. Thus, ironically, it is more essential to our well-being that we personally engage in leisure activities than that we personally engage in mandatory or maintenance activities.
One reason for the nebulous nature of the concept of leisure is that the boundaries between
leisure, mandatory, and maintenance activities can be quite permeable. This permeability occurs in three different ways – the first conceptually intrinsic to how the individual perceives an activity, the second largely facilitated by ICT, and the third often but not exclusively associated with ICT.

2.1.1 Permeable Boundaries (1): One Activity, Multiple Aspects
The first basis for the permeable boundaries between activity types is that intrinsically, many
activities possess characteristics of more than one of the conventional three categories (G`tz, et al., 2002; Meurs and Kalfs, 2000; Shaw, 1985; Tinsley, et al., 1993). This can be for a
combination of three different reasons: (1) The same activity may be experienced differently by different people; (2) the same activity may be experienced differently by the same person at
different times; and (3) an activity for a single person at a single time may mix aspects of
multiple categories.
Examples of the general principle come readily to mind: eating out or even cooking could be
considered maintenance activities, but are forms of recreation for many people. The same can be said of gardening and even housework or home repairs and improvements. Child care can be quite entertaining under the right circumstances (Shaw, 1984). Work-related travel and even commuting have some discretionary aspects for many (Mokhtarian et al., 2001; Redmond and Mokhtarian, 2001; Ory, et al., 2004). Hochschild (1997) points out that for many people, in contrast to the stereotype of the dog-eat-dog work world from which home is a serene refuge, work (where we interact with mature professionals who value our contributions) is a welcome escape from home (where we interact with needy and demanding family members). Howe and Rancourt (1990, p. 398) note that “[a] generally accepted theme of the psychology of leisure literature is that some people do find personal meaning and do experience freedom and leisure in work.” 7 And the recreational/ entertainment qualities of shopping (again, for some people) are well-recognized (Salomon and Koppelman, 1988; Tauber, 1972) 8. Even within the leisure category itself, an activity may have multiple characteristics. When one goes to a ball game with friends, is the activity social, or entertainment? The answer probably affects the activity choice process, including the choice set of perceived alternatives: if the primary motivation is social, one may first decide to get together with friends, and then choose an activity around which to organize the gathering, whereas if the primary motivation is entertainment, one may first decide to attend the ball game and then see who else is able to join.
This discussion speaks to the types and degrees of various motivations for undertaking a given
activity, which may differ from what the activity “label” itself would stereotypically imply (e.g. work is a necessary evil; leisure is an optional good). Understanding those motivations is
important for analyzing the leisure activity engagement decision process, and the role of ICT in that process. For example, Handy and Yantis (1997) hypothesize that the more chore-like the activity (i.e. the less that a mandatory or maintenance activity is viewed as having leisure
overtones), the greater the likelihood of in-home substitution for the out-of-home version of that activity.On the other hand, we are wary of endowing a mandatory or maintenance activity with leisure qualities simply because it can be pleasant. Meurs and Kalfs (2000) consider enjoyment to be an important element of the definition of leisure time, and it is tempting to equate enjoyment with leisure, suggesting that to the extent that mandatory or maintenance activities are enjoyed, they contain elements of leisure. But that may confuse the concepts of “positive utility” and leisure: a job can be enjoyable, stimulating, or fulfilling without being “leisurely”9. Conversely, not all leisure activities may be enjoyable: one may visit relatives but be miserable the entire time, or one may go to a gym in order to stay physically fit but consider it “torture”. We could say that a given activity constitutes leisure to people for whom it is enjoyable (see, e.g., the brief review of literature on “leisure as a state of mind” in Howe and Rancourt, 1990), whereas to those for whom it is not, it constitutes a form of maintenance – whether physical maintenance in the case of the gym, or social maintenance in the case of visiting family out of duty. But relying on subjective motivations as the basis for classifying the same activity differently for different people is not very practical for the large scale data collection and analysis needed for regional travel and activity modeling (although it may well be appropriate for more exploratory studies of activity and travel behavior, and as we discuss below, it is relevant for understanding activity choices in general and modeling ICT impacts on leisure travel in particular).

2.1.2 Permeable Boundaries (2): Multiple Types of Activities Fragmented and Sequentially Interleaved
Second, the boundaries between activity types are blurry due to what Couclelis (2000) refers to as the increasing fragmentation of activities, generally made possible by ICT. Whereas before, work, shopping, and leisure activities took place more or less in undivided blocks of time at specialized locations, we now see such activities broken into smaller chunks, interspersed with fragments of other activities, and spread across a larger number of locations. For example, we shop from the Internet or play computer games during a break at the office, and work from home in the evenings (perhaps interwoven with family interaction activities). We send and answer email while on vacation, and engage in sightseeing activities while on business trips (e.g., ECMT, 2000 points to the rise in “business tourism”) 10. This increasing fragmentability is also expected to have impacts on activity selection and scheduling, and the associated travel. For example, one may choose to watch a movie on DVD rather than in the theater precisely because the DVD can be stopped and started at will, and therefore woven into other activities at home rather than requiring the commitment of a larger block of time and a separate trip (although the travel involved in acquiring the DVD must still be taken into account, at least until downloading movies on demand becomes more widespread).

2.1.3 Permeable Boundaries (3): Multiple Types of Activities Simultaneously Overlapped

The third way in which boundaries between activity types are porous is simply due to
multitasking, a case in which fragments of multiple activities of different kinds actually
overlap 11. One may watch television (leisure) while doing a routine work task (mandatory) at home in the evening, or while cooking dinner (maintenance). One may phone a friend while
traveling home from work, make work-related calls while watching one’s child play soccer, or receive a call while eating with family or friends. Here again, the ability to multitask may affect one’s choice of activity mode, location, and timing.

2.1.4 Implications

The blurry boundaries between various leisure activities and between leisure and non-leisure
activities raise methodological complications. We have previously mentioned the impracticality of classifying the same activity as leisure or maintenance depending on one’s motivation for undertaking it or enjoyment of it. Data collection and analysis are also inherently complicated by the presence of fragmentation and multitasking among multiple activity types and subtypes within a short time period.
In sum, we are left with the sense that the more closely the concept of leisure is examined, the
more slippery it becomes. Although the considerations discussed above are important, as a
pragmatic (if somewhat unsatisfying) solution to the general question of defining leisure we may simply conclude, as US Supreme Court Justice Potter Stewart said about pornography, that we may not know how to define it, but we recognize it when we see it. Of course, empirical studies of leisure will ordinarily need to be more specific than this, and that can be accomplished by narrowing the definition for any particular investigation in ways that will best fit the objectives of that study (Samdahl, 1988).

2.2 Previous Classifications of Leisure Activities
Classification systems related to leisure activities and travel can be found in a number of
different contexts, including the literatures related to travel, activity analysis, time use, and
leisure. Although there are some interesting taxonomies based on the orientation of the individual toward leisure in general (Snir and Harpaz, 2002); personal values, personality, and lifestyle (Madrigal, 1995; Lanzendorf, 2002); or the purchase of leisure activities (Reid and Crompton, 1993), here we limit the discussion to studies that classify leisure activities themselves, according to various dimensions. At the simplest level, some typologies are based merely on the nature of the activity. For example, for the purposes of avocational counseling for the elderly, Overs, et al. (1977) classify activities under sports; nature; art and music; organizations; education, entertainment, and culture; volunteer; games; crafts; collecting. Passmore and French (2001) offer a simple tripartite classification: achievement leisure (playing sports, hobbies, creative and performance arts); social leisure (activities for the purpose of being in the company of others); and time-out leisure (listening to music, watching TV, contemplation).

Another relatively simple classification is based solely on purpose. For example, the 2001
National Household Transportation Survey uses two categories of trip purposes that could be
considered “leisure”: “social recreation” and “eat meal.” The social recreation category comprises five subcategories:
• go to gym/exercise/play sports,
• rest or relaxation/vacation,
• visit friends/relatives,
• go out/hang out (entertainment/theater/ sports event/go to bar),
• visit public place (historical site/museum/park/library).

The eat meal category comprises two subcategories:
• get/eat meal and
• coffee/ice cream/snacks.

Other typologies involve objective characteristics of the activity itself. For example, in addition to distinguishing social from recreational purposes, Meurs and Kalfs (2000) consider the dimensions of :
• number of overnight stays (day trips, short stays of 1-3 nights, short holidays of 3-5 nights
away, and long holidays of more than 5 nights away);
• trip length (short trips of up to two hours, and day journeys of more than two hours);
• destination location type (local, regional, national, international); and
• role of journey (purely to reach a destination, versus having an intrinsic recreational value);

where the latter dimension of role is subjective rather than objective. Bhat and Lockwood (2003) classify weekend out-of-home social/recreational activities according to whether they are physically active or passive, and whether they constitute travel itself (e.g. jogging, cycling) or take place at a specific out-of-home location.
Several classifications of leisure activities are based primarily or in part on individual values or psychological needs. For example, Holmberg, et al. (1990) list 760 leisure activities classified by combinations of two of the following six interest dimensions: realistic, investigative, artistic, social, enterprising, conventional. Tinsley and Eldredge (1995) developed a taxonomy of leisure activities based on their psychological benefits. Starting with a list of 82 leisure activities and an empirical rating of each leisure activity for eleven different psychological benefits, they used cluster analysis to define 12 classes of leisure activities (Table 1). The psychological basis of these classes is appealing in that it might provide a convenient way of hypothesizing which kinds of leisure activities are more likely to be impacted by ICT and in what ways. For example, agency activities involve physical exertion that is not required for ICT-based activities.
Activities fulfilling the “novelty,” “belongingness,” and “sensual enjoyment” needs also seem
unlikely candidates for substitution (the category 1 effect of ICT discussed in Section 3.1 below).
For all of these activity classes, however, ICT may play an important role in managing travel and may even generate travel (the category 4 effect). Activities fulfilling other needs, such as
cognitive simulation, self-expression, and creativity, do not so clearly necessitate travel to begin with, in which case ICT may provide a new dimension to the participation in these activities (the category 2 effect).

Continua >>>>>

4 Anable (2002, p. 181) comments that leisure “represents one of the only journey purposes with essentially universal participation”, and G`tz (2003) found that there was less variability across lifestyle clusters in the time devoted to leisure activities than in the time spent on non-leisure.
5 The exceptions are those maintenance activities that must be performed directly on/by the individual herself, such as eating, personal grooming, and medical appointments.
6 Again, there are exceptions: some leisure activities undertaken out of duty to other people (see discussion below) may occasionally be outsourced, as when we get someone to take our place at a social or entertainment event we really do not wish to attend.
7 For similar views on the social-psychological fulfillment aspects of work, see Csikszentmihalyi and LeFevre (1989) and Tschan, et al. (2004); see Lewis (2003) for a thoughtful and balanced discussion of whether professional knowledge work is “the new leisure”. For a divergent perspective, in which “exciting and strenuous” leisure pursuits are chosen in deliberate contrast to “boring and sedentary” jobs, see Kernan and Domzal (2000, p. 97).
8 It is perhaps not coincidental that all the examples just given involve a location-based version of the activity rather than an ICT-based version. It may well be that the leisure aspects of a mandatory or maintenance activity are stronger in its location-based form, although on-line shopping seems to have a strong leisure component.
9 For example, a high-stress occupation such as stockbroker may be all of those things (much of the time) without being considered leisurely. On the other hand, the opposite condition, relaxation, cannot be used to define leisure, since many leisure activities such as those involving strenuous physical exercise would not be considered relaxing.
10 Whether constantly being “on call” is a desirable condition is of course debatable, and probably differently
desirable for different people. Our point is simply that it is a reality for many people, with real implications for travel.
11 The boundary between this category and the preceding one is also blurry, technically depending on whether the interspersed activity fragments occur one at a time, or overlap. In practice it can be difficult to make this distinction, depending in part on the time scale at which activities are distinguished. A 10-minute Internet shopping episode at work could be distinguished separately (constituting sequential interleaving) if the time scale were in minutes, but would be considered multitasking (a secondary activity overlapping the primary activity of work) if the time scale were in hours.